31–40 of 42 posts
That problem is the AI problem these game designers are trying to tackle.
Just seems interesting...
Earlier quoted context omitted.
I'd noticed too that the winner claimed to use just a variety of hand-rolled strategies, and wonder why that is. Has anyone had the time to dig into the code? Is this a case of well-writen "naive" code trumping theoretically-better but poorly-implemented algorithms, or are there properties of the game state-space which make e.g. Monte Carlo tree search perform poorly?
Speculation: the winning author is simply good at StarCraft and knew what to implement. His implementation was probably simple but did one thing very well, which beat more complicated/smart AI implementations from worse players. Monte Carlo over the StarCraft state-space would be very hard to implement but I don't see any particular reason why it wouldn't work decently well.
Would be interesting to get your hands on a large set of replays from a lot of gosu players and run some machine learning offline prior to each map for the current match map. Would be fun to experiment with the data and see if it would be possible to infer game tactics from professional players and "simulate" those against another bot based on the current game state.
There are probably too many variables in Starcraft for this to be practical. With a game like chess there's a fairly obvious set of state variables to keep track of (max of 32 pieces that each have a position in essentially an 8x8 2D array), with Starcraft it's pretty nuts though, each side can have dozens of buildings, dozens of units of ~15 types, each of which has a position on a much larger grid, health, sometimes energy, status effects, etc. Obviously there are RTS games that are more complicated yet, but the jump from a turn-based board game to an RTS is just huge.
Really interesting, I started playing on SC2 so I'm not sure about the dynamics of BW but I wonder how the lack of micro skews the results. For example a fast reaper push hopping on/off the enemy base can't probably be done by a bot. Terrans are also much easier to automate, while Zergs require microing with injections, etc are the demos uploaded anywhere?
That's really cool. Is there a tldr of the AI strategies used ? Are some of the bots using neural network?
1. Human players use scouts to counter fog of war followed by looking at build patterns to determine likely strategy. Spotting strategy should be easy for pattern matching. They also often micro the scout to (a) disrupt mining or (b) keep it alive for as long as possible. This should be similarly easy.
Note: Units damaged from a prior battle can also be used as scouts and simply not factored into further combat.
2. Simulations should be done using most likely combos of units for each strategy against most likely combos of opponents' units. This should be done for macro and micro. Patterns here will be used in the micro-engine. Bonus against machines: use unlikely combinations that prove effective in simulations.
3. Combat analysis should be a global thing that factors into overall goals. That the human player moved units from one base to another should never surprise an AI. The AI should've shifted some or all defence to that base immediately with use of fast-moving scouts (eg zerglings) to track opponents if they run.
4. Building laying and build order heuristics should be datamined from top players while attached to specific strategies. No one size fits all. Have two for each: one ideal, optimized build order; a heuristic, on-the-fly set if an attack or change throws off the plan.
5. Identify the common strategy changes that happen as a result of what enemy does. Simulate likely situations with units or build orders for those. Identify heuristics for safe transition. Will help for adaptation.
6. Put as many human strategies in there as possible from the A-listers, both specifically and generically.
Those are where I'd start on improving the situation, esp against human players. However, I think an even more interesting line of research is augmenting human players with an AI to efficiently execute their strategy. One that constantly assesses, micro's, follows build orders, etc. Basically never misses a step. More practical for real-world applications.
Really interesting, I started playing on SC2 so I'm not sure about the dynamics of BW but I wonder how the lack of micro skews the results. For example a fast reaper push hopping on/off the enemy base can't probably be done by a bot. Terrans are also much easier to automate, while Zergs require microing with injections, etc are the demos uploaded anywhere?
This couldn't be more wrong. The amount of glitches in BW that allow got micro potential is huge. It's actually one of the biggest complaints from BW players about SC2, the lack of available micro. Check out th video "Depth of Micro" by Lalush for more info on things like muta stacking, patrol micro, carrier leash range, and tons of other hidden micro. You are also wrong about your last statement, the limiting factor…
Slightly off-topic but I find the BWAPI more interesting than the AI aspects. Does anyone have any further reading regarding this kind of thing?
Would be interesting to get your hands on a large set of replays from a lot of gosu players and run some machine learning offline prior to each map for the current match map. Would be fun to experiment with the data and see if it would be possible to infer game tactics from professional players and "simulate" those against another bot based on the current game state.